Grifoni, Alba, Pham, John, Sidney, John, O'Rourke, Patrick H, Paul, Sinu, Peters, Bjoern, Martini, Sheridan R, de Silva, Aruna D, Ricciardi, Michael J, Magnani, Diogo M, Silveira, Cassia GT, Maestri, Alvino, Costa, Priscilla R, de-Oliveira-Pinto, Luzia Maria, de Azeredo, Elzinandes Leal, Damasco, Paulo Vieira, Phillips, Elizabeth, Mallal, Simon, de Silva, Aravinda M, Collins, Matthew, Durbin, Anna, Diehl, Sean A, Cerpas, Cristhiam, Balmaseda, Angel, Kuan, Guillermina, Coloma, Josefina, Harris, Eva, Crowe, James E, Stone, Mars, Norris, Phillip J, Busch, Michael, Vivanco-Cid, Hector, Cox, Josephine, Graham, Barney S, Ledgerwood, Julie E, Turtle, Lance, Solomon, Tom, Kallas, Esper G, Watkins, David I, Weiskopf, Daniela, and Sette, Alessandro
While progress has been made in characterizing humoral immunity to Zika virus (ZIKV) in humans, little is known regarding the corresponding T cell responses to ZIKV. Here, we investigate the kinetics and viral epitopes targeted by T cells responding to ZIKV and address the critical question of whether preexisting dengue virus (DENV) T cell immunity modulates these responses. We find that memory T cell responses elicited by prior infection with DENV or vaccination with tetravalent dengue attenuated vaccines (TDLAV) recognize ZIKV-derived peptides. This cross-reactivity is explained by the sequence similarity of the two viruses, as the ZIKV peptides recognized by DENV-elicited memory T cells are identical or highly conserved in DENV and ZIKV. DENV exposure prior to ZIKV infection also influences the timing and magnitude of the T cell response. ZIKV-reactive T cells in the acute phase of infection are detected earlier and in greater magnitude in DENV-immune patients. Conversely, the frequency of ZIKV-reactive T cells continues to rise in the convalescent phase in DENV-naive donors but declines in DENV-preexposed donors, compatible with more efficient control of ZIKV replication and/or clearance of ZIKV antigen. The quality of responses is also influenced by previous DENV exposure, and ZIKV-specific CD8 T cells from DENV-preexposed donors selectively upregulated granzyme B and PD1, unlike DENV-naive donors. Finally, we discovered that ZIKV structural proteins (E, prM, and C) are major targets of both the CD4 and CD8 T cell responses, whereas DENV T cell epitopes are found primarily in nonstructural proteins.IMPORTANCE The issue of potential ZIKV and DENV cross-reactivity and how preexisting DENV T cell immunity modulates Zika T cell responses is of great relevance, as the two viruses often cocirculate and Zika virus has been spreading in geographical regions where DENV is endemic or hyperendemic. Our data show that memory T cell responses elicited by prior infection with DENV recognize ZIKV-derived peptides and that DENV exposure prior to ZIKV infection influences the timing, magnitude, and quality of the T cell response. Additionally, we show that ZIKV-specific responses target different proteins than DENV-specific responses, pointing toward important implications for vaccine design against this global threat.